High-Throughput Preosteoblastic Spheroids Elevate Fibroblast Growth Factor 23 via Parathyroid Hormone Signaling Pathway

Abstract

Fibroblast growth factor 23 (FGF23) plays a crucial role in managing renal phosphate and the synthesis of 1,25(OH)2-vitamin D3, which is essential for bone homeostasis. Developing robust in vitro systems to study FGF23-regulating mechanisms is crucial for advancing our knowledge and identifying potential therapeutic targets. The traditional in vitro 2D culture system results in relatively low expression of FGF23, complicating further exploration of its regulatory mechanisms and potential therapeutic targets. Herein, we reported a high-throughput approach to generate preosteoblastic cell spheroids with enhanced FGF23 production. For this purpose, murine preosteoblast cell line (MC3T3-E1) was cultured in our previously reported nonadherent microwells (200 µm in diameter, 148 µm in depth, and 100 µm space in between) and self-assembled into spheroids with a diameter of 92.3 ± 15.0 µm after 24 h. Compared with monolayer culture, the MC3T3-E1 spheroids showed a significant upregulation of FGF23 in both gene and protein levels after 24 h of serum-free induction. RNA sequencing and western blotting analysis further suggested that the enhanced FGF23 production in MC3T3-E1 spheroids was attributed to the activation of the parathyroid hormone (PTH)/PTH1R signaling pathway. Impressively, inhibition of PTH signaling through small molecular inhibitors or short hairpin RNA targeting PTH1R effectively reduced FGF23 production. In summary, the current study revealed the efficacy of the high-throughput formation of preosteoblast cell spheroid in stimulating FGF23 expression for mechanistic studies. Importantly, our findings highlight the potential of the current 3D spheroid system for target identification and drug discovery.

Impact Statement

This research introduces a high-throughput method for generating preosteoblastic spheroids that significantly enhance fibroblast growth factor 23 (FGF23) production via the parathyroid hormone signaling pathway. The findings offer a robust in vitro model for studying FGF23 regulation, addressing the limitations of traditional 2D cultures. This advancement not only improves our understanding of bone and mineral metabolism but also holds potential for identifying novel therapeutic targets for conditions with elevated FGF23 levels, such as chronic kidney disease and metabolic bone disorders. The scalable and efficient spheroid formation method could revolutionize the approach to drug discovery and target validation for managing FGF23 production.

Introduction

Fibroblast growth factor 23 (FGF23), a critical hormone primarily produced by osteoblasts and osteocytes in the bone, is essential in managing phosphorus, calcium, and vitamin D homeostasis.¹ It regulates phosphate reabsorption in the kidneys and inhibits the production of active vitamin D, thus maintaining the balance of phosphate and calcium in the body.^2,3 Elevated levels of FGF23 are associated with multiple pathological conditions, such as chronic kidney disease (CKD), tumor-induced osteomalacia, and certain genetic disorders such as hypophosphatemic rickets.^4–6 The pathological consequences of dysregulated FGF23 highlight the necessity of understanding the regulatory mechanisms governing its production. Developing robust in vitro systems to study FGF23-regulating mechanisms is crucial for advancing our knowledge and identifying potential therapeutic targets.

Traditional in vitro studies of FGF23 typically use 2D culture systems with osteocytic and osteoblastic cell lines, such as MC3T3-E1 and MLO-Y4 cells.^7–9 While 2D cultures are relatively simple and cost-effective, they often result in very low levels of FGF23 expression and a limited response to stimulants, which does not accurately reflect the physiological conditions in vivo.^10,11 One possible reason is that cells cultured in a 2D monolayer exhibited diminished mechanosensory and mechanotransduction capabilities, resulting in reduced mechanical stimuli that promote FGF23 synthesis.¹² Moreover, 2D cultures altered cell metabolic activity compared with the in vivo physiological state, which also affects the synthesis and section of FGF23.^13,14 Furthermore, 2D cultures lack the hierarchical structure and cell–cell contact that are critical for the function of osteocytes and osteoblasts and FGF23 production.¹⁵ Consequently, the regulatory mechanisms and potential therapeutic targets related to FGF23 cannot be effectively explored in 2D cultures.

To address the limitations of 2D cultures, researchers have explored cell-laden biomaterials that are archetypically seeded with individual cells and steered into the desired behaviors using exogenous stimuli to control growth and differentiation.¹⁶ A previous study showed that the osteoblasts and osteocytes laden in Collagen I hydrogel exhibit elevated in vitro FGF23 production.¹⁷ Furthermore, the encapsulated cells also gain a more stellate-like morphology and demonstrate a more robust response to parathyroid hormone (PTH) treatment and mechanical loading.¹⁸ One study found that growing cells under 3D models produced more realistic mechanical effects and showed significant differences in FGF23 expression.¹⁹ In addition, previous studies have also used a 3D bone tissue model to investigate FGF23 expression levels in bone cells, showing that 3D models more accurately reflect FGF23 expression levels.²⁰ However, the encapsulation of individual cells in biomaterials often creates the vast cell-free areas within the biomaterial, thereby lacking the direct cell–cell interactions necessary for the maturation of osteoblasts and osteocytes.^21,22

In order to mimic cell–cell interactions, several cellular spheroids culture approaches, including centrifugation, low-adherence substrate, and hanging drop culture, have been proposed.^23,24 However, the aforementioned approaches are not only laborious but also difficult for in vivo applications due to scalability challenges and uncontrolled complexity in 3D cell culture formats.²⁵ Furthermore, the underlying mechanism regulating the differentiation and maturation of cellular spheroids remains elusive.

We previously developed a high-throughput platform, which allows individual cells to self-aggregate into cellular spheroids containing 50–250 cells.^16,26 Our previous work demonstrates the facile production of high quantities of skeletal progenitors, which exhibited enhanced differentiation capacities compared with the 2D monolayer counterparts.²⁶ Herein, we report the efficacy of the high-throughput formation of osteoblastic spheroids in stimulating FGF23 expression, which involves activation of PTH/PTH1R signaling through several key components, including the phosphorylation of protein kinase A (PKA) and cAMP response element-binding protein (CREB). Our findings highlight the potential of this 3D spheroid system in target identification and drug discovery for managing FGF23 production.

Materials and Methods

Cell culture

Mouse preosteoblast cell line (MC3T3-E1) was cultured at 37°C with 5% carbon dioxide (CO₂) in alpha modification of Eagle’s medium (α-MEM) (HyClone, SH30265.01) supplemented with 10% fetal bovine serum (FBS; Gibco, No.2324371) and 1% penicillin–streptomycin (Cytiva, SV30010). When cell confluence reached 80%, cells were detached and 2 × 10⁵/well seeded into plates. Cells were then incubated in α-MEM conditions for 24 h. For inhibitory experiments, MC3T3-E1 cells were pretreated with 10 µM H89²⁷ (Beyotime, S1643), diluted in dimethyl sulfoxide (DMSO) (PanReac AppliChem, A3672) to a final DMSO concentration of 0.1%. This pretreatment occurred 1 h before cell seeding. Subsequently, the cells were seeded and exposed to 10 µM H89 for 24 h. Controls were in 0.1% DMSO for 24 h. Cells were harvested for analysis after 24 h. For all in vitro experiments, unless otherwise specified, triplicates (n = 3) were used in each condition, and all the in vitro experiments were repeated for two runs.

High-throughput formation of MC3T3-E1spheroids

The expanded MC3T3-E1 was seeded in our previously reported agarose 24-well inserts containing ≈2000 microwells (Leijten et al.¹⁶) at a concentration of 400,000 cells/well. Cells were cultured in serum-free medium at 37°C and 5% CO₂ for 24 h. After that, the obtained spheroids were harvested by simple pipetting and centrifuged at 1000 g for 5 min for further analyses.

Live/dead staining

After 24 h of culture, cells cultured in monolayers (2D) and spheroids (3D) were removed from the culture medium and rinsed three times with phosphate-buffered saline (PBS, pH = 7.4). Then, live/dead staining was performed to assess the viability using Beyotime C2015S according to the manufacture’s instruction. Fluorescent images of live and dead cells were captured using a confocal microscope (Leica SP8; Leica) with excitation light of 490 and 545 nm, respectively. Images were analyzed using ImageJ software (ver. 1.51j).

Immunocytochemistry

Cell spheroids were collected from the culture plates and fixed with 4% paraformaldehyde. Thereafter, cells were permeabilized with 0.5% Triton X-100 in PBS and blocked with PBS containing 5% FBS and 0.1% Triton X-100. Next, Alexa Fluor 594-conjugated Phalloidin (UE, 1:100) in the blocking buffer for 2 h after three washes counterstained with Hoechst 33342 (Beyotime, 1:1000). Spheroids were then mounted with antifade (Invitrogen). Monolayer cells were grown on coverslips and stained using the same dilution ratio and protocol.

RNA sequencing and data analysis

Total RNA was extracted from spheroids and monolayer-cultured cells using the EZNA HP Total 19 RNA Kit (Omega). RNA quality was examined using an Agilent 2100 bioanalyzer. The bulk RNA sequencing (RNAseq) was conducted by the Beijing Genomics Institute (BGI). Single-stranded circular DNA libraries were prepared through the processes of polyadenylated RNA enrichment, reverse transcription, bubble adapter ligation, and polymerase chain reaction (PCR) amplification. Libraries were sequenced on the DNBSEQ platform by BGI. Raw data were filtered with SOAPnuke software to generate clean reads and mapped to the murine genome (GRCm38.p6) using HISAT software. Clean reads were then aligned to reference gene sequences using Bowtie.²⁸ Subsequently, the DESeq2 (v1.4.5) (www.bioconductor.org/packages/devel/bioc/html/DESeq2.html) software was used for data normalization and identifying differentially expressed genes between spheroids (3D) and monolayer-cultured cells (2D) with a Q value (adjusted p value) cutoff of 0.05. The volcano plot was created using the TBtools.²⁹ Kyoto Encyclopedia of Genes and Genomes (http://www.genome.jp/kegg/, accessed on January 1, 2022) pathway analyses were performed using the RNA Data Visualization System Dr. TOM system (BGI; http://www.bgi.com/global/dr-tom/).

Quantitative real-time PCR

The total RNA of cells was extracted with the EZNA HP Total RNA Kit (Omega) and reverse transcribed using HiScript III All-in-One RT SuperMix (Vazyme#R222). Quantitative reverse transcription PCR (qRT-PCR) was performed using the HiScript II One Step qRT-PCR SYBR Green Kit (Vazyme#Q711) on the CFX Connect Real-Time System (Bio-Rad#1855201). Primers used in this study were listed in Table 1, and the primers were synthesized by Aoke Biotech Co, Ltd. The sample size is five for each gene expression in each condition (n = 5). The expression of genes was quantified using Gapdh as a housekeeping gene, and the expressions of each gene were calculated using the 2^−ΔΔCt method.

Table 1.

The Primers Used in This Study

Gene	Sequence (5′−3′)
Fgf23	Forward	ATGCTAGGGACCTGCCTTAGA
Fgf23	Reverse	AGCCAAGCAATGGGGAAGTG
Fos	Forward	CGGGTTTCAACGCCGACTA
Fos	Reverse	TTGGCACTAGAGACGGACAGA
Egr1	Forward	TCGGCTCCTTTCCTCACTCA
Egr1	Reverse	CTCATAGGGTTGTTCGCTCGG
Mef2c	Forward	ATCCCGATGCAGACGATTCAG
Mef2c	Reverse	AACAGCACACAATCTTTGCCT
Ocn	Forward	GCAATAAGGTAGTGAACAGACTCC
Ocn	Reverse	GCGTTTGTAGGCGGTCTTCAAG
Pth1r	Forward	GTACACCGAGGTCTCAGGGA
Pth1r	Reverse	CGCTTGAAGTCCAATGCCAG
Rankl	Forward	CAAGCTCCGAGCTGGTGAAG
Rankl	Reverse	CCTGAACTTTGAAAGCCCCA
Dmp1	Forward	CGCATCCCAATATGAAGACTG
Dmp1	Reverse	GCTTGACTTTCTTCTGATGACTCA
Sost	Forward	GCCTCATCTGCCTACTTGTG
Sost	Reverse	CTGTGGCATCATTCCTGAAG
Gapdh	Forward	TGTGTCCGTCGTGGATCTGA
Gapdh	Reverse	TTGCTGTTGAAGTCGCAGGAG
shPthr1	Forward	CCGGTGGACGCAGACGATGTCTTTACTCGAGTAAAGACATCGTCTGCGTCCATTTTTG
shPthr1	Reverse	AATTCAAAAATGGACGCAGACGATGTCTTTACTCGAGTAAAGACATCGTCTGCGTCCA

Western blotting

Cells were lysed on ice in a lysis buffer (Merck Millipore) with a proteinase inhibitor (MedChemExpress, HY-K0010) and processed by centrifugation at 13,000 rpm for 10 min at 4°C. Protein concentrations were measured using a BCA Protein Assay Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. For each sample, 15 µg of denatured protein per well was loaded onto a 10% polyacrylamide gel, separated by electrophoresis, and then transferred onto nitrocellulose membranes (Millipore). After blocking with 5% nonfat milk in TBST for 1 h at room temperature, membranes were incubated overnight at 4°C with the following primary antibodies: rabbit anti-Phospho-PKA Substrate (1:1000; Cell Signaling Technology#9624S), rabbit anti-Phospho-creb1 (1:1000; Abclonal#AP0019), rabbit anticreb1 (1:1000; Abclonal#A11989), goat anti-FGF23 (1:4000; Immunotopic #186–206), and mouse anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (1:8000; Proteintech#60004-1-I). On the next day, membranes were incubated with horseradish peroxidase-conjugated secondary antibody (Goat anti-Rabbit IgGAS014, 1:4000; or Goat anti-Mouse, AS071,1:4000) diluted in 5% nonfat milk for 1 h at room temperature. Signals were detected using WesternBright ECL solution (Advansta), and densitometry analysis that was standardized to GAPDH was performed with Fiji software. At least, triplicates (n = 3) were used for western blotting quantification.

Plasmid construction

In this study, we leveraged the GPP WEB PORTAL (GPP Web Portal—The TRC shRNA Design Process [broadinstitute.org]) for the design of shRNA sequences, prioritizing the ones with high predicted frequencies. The primers used in this study to construct the hairpin RNA (shRNA) targeting Pth1r are listed in Table 1. The shRNA sequences were cloned into the pLKO.1 puro vector (Addgene, #8453) after digestion with AgeI and EcoRI. The final construct, pLKO.1 shPth1r, was validated by sequencing to confirm the positive insertion of the shPth1r sequences.

Lentivirus production

HEK293T cells (Pricella, CL-0005) were seeded in 6-well plates at a density of 6 x10⁵ cells per well and grown overnight. The following day, the cells were gently washed using PBS, followed by the addition of 0.5 mL of Dulbecco’s modified Eagle’s medium (DMEM) (Cytiva, SH30262.01) without FBS was added into each well. Simultaneously, a transfection mixture was prepared with 900 ng of psPAX2 (Addgene, 12260), 300 ng of pMD2.G (Addgene, 12259), and 1200 ng of shPth1r, along with 6 µL of Lipofectamine 2000 (Lipofectamine™, 11668027). This mixture was further diluted in 0.15 mL of OptiMEMTM (Gibco) and added to each well. The cells were incubated at 37°C for 8 h, after which 0.5 mL DMEM supplemented with 10% (v/v) FBS was replenished. The medium containing lentiviruses was harvested 48 h after transfection.

Lentiviral transduction of MC3T3-E1 cells

MC3T3-E1 cells were seeded in 6-well plates at a density of 1.5 × 10⁴ cells/cm². When the cells reached 70% confluence, they were transfected with either the experimental lentivirus (shPth1r) or the negative control (pLKO.1 shCon) lentivirus at 37°C for 8 h. Cells were selected with a selective medium containing 5 µg/mL puromycin (Beyotime, ST551) for 72 h to establish stable knockdown clones. After 4 days, the t stable clones were collected for subsequent experiments. The knockdown efficiency of Pth1r was confirmed using reverse transcription-quantitative PCR. Triplicates (n = 3) were used for the lentiviral transduction experiment.

Statistical analysis

All the numerical data were represented as mean ± standard deviation, and data were plotted as individual data points with bars representing the average value. Statistical analysis was performed using the GraphPad Prism software. Shapiro–Wilk test was used to test for normal distribution. A nonpaired unequal variance Student’s t-test or Mann–Whitney test was used for two-group experiments. For three-group comparisons, a one-way analysis of variance followed by a Bonferroni multiple comparisons posttest was applied. Statistical significance is indicated on all tables and graphs as follows: *p < 0.05. **p < 0.01, ***p < 0.001.

Results

High-throughput formation of osteogenic progenitor spheroids with elevated FGF23 production

The murine calvaria preosteoblast cell line MC3T3-E1 was seeded onto a 24-well plate containing microwells with 200 µm in diameter, 148 µm in depth, and 100 µm space in between (Fig. 1A). After 3 h, 200 cells were self-assembled in each microwell and formed a spheroid with a diameter of 92.3 ± 15.0 µm (Fig. 1B, C). The obtained spheroids remained stable during 24 h in vitro culture, and the average diameter decreased to 51.4 ± 5.8 µm after 24 h (Fig. 1B, C).

FIG. 1.

The high-throughput generation of murine osteoblastic progenitors (MC3T3-E1) spheroid enhanced osteoblastic differentiation and FGF23 production. (A) The diagrams of the process of MC3T3-E1 spheroids formation. (B) Light images of the morphology and self-assembly process of MC3T3-E1 spheroids over 24 h in vitro culture in the nonadherent agarose microwell array. Scale bar = 200 μm. (C) Size quantification of in vitro cultured MC3T3-E1 spheroids at 3, 12, and 24 h, respectively. Over 30 spheroids from 3 random fields in 3 different wells were counted for each time point. One-way analysis of variance was used for the statistical analysis. (D) Live/dead staining and immunofluorescent staining of Hoechst 33342 (blue) and phalloidin (red) of MC3T3-E1 spheroids (3D) or MC3T3-E1 monolayer (2D) at 24 h. (E) Relative mRNA expression of Fgf23 in MC3T3-E1 cells cultured in monolayer (2D) and spheroids (3D) after 24 h (n = 5). (F) Western blotting analysis of FGF23 expression of MC3T3-E1 in monolayer (2D) or spheroids (3D) at 24 h (n = 5). (G) Densitometry of FGF23 normalized to GAPDH. (H–J) Relative mRNA expression of the osteogenic genes Ocn (H), Dmp (I), and Sost (J) in MC3T3-E1 cells cultured in monolayer (2D) or spheroids (3D) after 24 h (n = 3). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Live and dead staining revealed that the majority of cells within the spheroids remained viable after 24 h (Fig. 1D). Compared with monolayer cells, the MC3T3-E1 spheroids exhibited distinct cytoskeleton reorganization over 24 h in vitro culture (Fig. 1D). Furthermore, MC3T3-E1 spheroids exhibited a significantly higher expression of FGF23 than monolayer cells, with approximately a 10-fold increase in mRNA levels and a 2-fold increase in protein levels (Fig. 1E–G). Additionally, several osteogenic markers, including Ocn, Dmp1, and Sost, were also significantly upregulated (Fig. 1H–J).

The osteogenic progenitor spheroids enhanced PTH/PTH1R signaling pathway

To better explore the mechanism underlying the elevated FGF23 production of the osteogenic progenitor spheroids, we conducted RNAseq analysis of monolayer cultured (2D) and spheroids (3D) of MC3T3-E1. Compared with the monolayer (2D), MC3T3-E1 spheroids displayed dramatic gene expression changes, with a total of 1160 upregulated genes [ $\log_{2}$ FC(3D/2D) > 1, Q value <0.05] and 1213 downregulated genes [ $\log_{2}$ FC(3D/2D) < −1, Q value <0.05] within 24 h (Fig. 2A). Gene ontology enrichment analysis with upregulated genes revealed significant enrichment in cell differentiation, response to hormone, and cell surface receptor signaling pathway (Fig. 2B). Quantitative PCR data revealed a significant upregulation of PTH signaling downstream targets such as Fos, Egr1, Mef2c, and Rankl specifically in MC3T3-E1 spheroids, indicating the activation of PTH signaling (Fig. 2C–F).

FIG. 2.

Murine osteoblastic progenitors (MC3T3-E1) MC3T3-E1 spheroids enhanced FGF23 expression via PTH signaling. (A) Volcano plot of the differentially expressed genes generated from bulk RNA sequencing of MC3T3-E1 cells cultured in monolayer (2D) or spheroids (3D) at 24 h. Q value <0.05, log₂ (3D/2D) > 1 was defined as upregulated (red dots), whereas Q value <0.05, log₂ (3D/2D) < −1 was defined as downregulated (blue dots). (B) Gene ontology enrichment analysis of the biological process of the upregulated genes in spheroids. (C–F) The real-time qPCR verification of the upregulated genes Fos (C), Egr1 (D), Mef2c (E), and Rankl (F) of MC3T3-E1 cells cultured in monolayer (2D) or spheroids (3D) at 24 h in the PTH pathway (n = 3). (G) Western blotting analysis of phosphorylated protein kinase A (p-PKA), phosphorylated cyclic adenosine monophosphate (cAMP) response element-binding protein (p-CREB), and total CREB of MC3T3-E1 in monolayer (2D) or spheroids (3D) at 24 h (n = 4). (H) Densitometry of phosphorylated PKA normalized to GAPDH (n = 4). (I) Densitometry of phosphorylated CREB normalized to total CREB (n = 4). *p < 0.05; **p < 0.01; ***p < 0.001.

To verify the activation of the PTH signaling pathway in MC3T3-E1 spheroids, we performed western blotting on MC3T3-E1 spheroids and monolayer cells over 24 h in vitro culture in serum-free condition. The spheroids (3D) showed significantly higher levels of phosphorylated PKA (p-PKA) and CREB (p-CREB) than monolayer-cultured cells (2D) (Fig. 2G-I), suggesting that the FGF23 upregulation observed in spheroids was potentially regulated by the activation of the PTH/PTH1R signaling pathway.

Inhibition of PTH/PTH1R signaling attenuated FGF23 expression in MC3T3-E1 spheroids

To further verify whether PTH/PTH1R signaling mediates FGF23 expression in MC3T3-E1 spheroids, we knocked down PTH receptor 1 in MC3T3-E1 cells using lentiviral sh-Pth1r (Fig. 3A). Quantitative PCR data confirmed the sh-RNA mediated silencing of PTH1R was associated with an ∼60% reduction of Pth1r mRNA (Fig. 3B). Compared with the control group, the knockdown of Pth1r impaired the PTH/PTH1R signaling, evidenced by a significant decrease of p-PKA and p-CREB (Fig. 3C–E), in line with a significant downregulation of Fos, the PTH/PTH1R signaling target gene (Fig. 3F). Consequently, the sh-Pth1r group showed a decline in FGF23 expression in both mRNA (Fig. 3G) and protein levels (Fig. 3H-I).

FIG. 3.

Knockdown of Pth1r attenuated FGF23 expression in murine osteoblastic progenitors MC3T3-E1 spheroids. (A) Schematic representation of suspension transfection of MC3T3-E1 spheroids with the shPth1r lentivirus. (B) Relative mRNA expression levels of Pth1r in MC3T3-E1 spheroids cultured in spheroids at 24 h in control or Pth1r knockdown cells (n = 3). (C) Western blotting results of phosphorylated PKA, phosphorylated CREB, and total CREB protein expression levels in MC3T3-E1 spheroids cultured in spheroids at 24 h in control or Pth1r knockdown cells (n = 3). (D) Densitometry of phosphorylated protein kinase A (p-PKA) normalized to GAPDH. (E) Densitometry of phosphorylated cAMP response element-binding protein (p-CREB) normalized to total CREB (n = 3). (F) Relative mRNA expression levels of Fos in MC3T3-E1 spheroids cultured in spheroids at 24 h in control or Pth1r knockdown cells (n = 3). (G) Relative mRNA expression levels of Fgf23 in MC3T3-E1 spheroids cultured in spheroids at 24 h in control or Pth1r knockdown cells (n = 3). (H) Western blotting results of FGF23 protein expression in MC3T3-E1 spheroids cultured in spheroids at 24 h in control or Pth1r knockdown cells. (I) Densitometry of FGF23 normalized to GAPDH (n = 3). *p < 0.05; **p < 0.01.

Next, we conducted a PKA kinase activity inhibition experiment to verify the influence of PTH/PTH1R signaling on regulating FGF23 expression in MC3T3-E1 spheroids. To this end, MC3T3-E1 in spheroids was cultured in serum-free conditions for 24 h and treated with or without the PKA inhibitor, H-89 (10 μM). Upon inhibition, MC3T3-E1 spheroids showed a significant decrease of p-PKA (up to 40%), although the phosphorylation of CREB remains unaffected (Fig. 4A–C). Consistently, the FGF23 expression in both mRNA and protein levels was also evidently downregulated in MC3T3-E1 spheroids upon inhibition of PKA kinase activity (Fig. 4D–F). These results demonstrate the regulatory effect of the PTH/PTH1R signaling pathway on FGF23 expression in MC3T3E1 spheroids (Fig. 5).

FIG. 4.

Inhibition of protein kinase A (PKA) kinase activity attenuated FGF23 expression in murine osteoblastic progenitors (MC3T3-E1) spheroids. (A) Western blotting results of phosphorylated PKA, phosphorylated CREB, and total CREB protein expression levels after 24 h of treatment with or without the PKA inhibitor (10 μM H-89) (n = 3). (B) Densitometry of p-PKA normalized to GAPDH (n = 3). (C) Densitometry of phosphorylated CREB normalized to total CREB (n = 3). (D) Relative mRNA expression levels of Fgf23 in MC3T3-E1 spheroids cultured in spheroids after 24 h of treatment with or without the PKA inhibitor (10 μM H-89) (n = 3). (E) Western blotting results of FGF23 protein expression in MC3T3-E1 spheroids after 24 h of treatment with or without the PKA inhibitor (10 μM H-89) (n = 3). (F) Densitometry of FGF23 normalized to GAPDH (n = 3). *p < 0.05.

FIG. 5.

Schematic representation of the activation of PTHR1/PKA/CREB signaling elevated FGF expression in murine osteoblastic progenitors (MC3T3-E1) spheroids. Self-assembly of individual MC3T3-E1 cells into spheroids facilitates the activation of cell surface receptor PTH1R, subsequently leading to the activation of protein kinase A (PKA). Consequently, phosphorylated PKA increases the phosphorylation of cAMP response element-binding protein (CREB), and the transcription of osteogenic differentiation marker genes Ocn, Dmp1, and Sost, thereby upregulating Fgf23 expression. AC, adenylyl cyclase.

Discussion

FGF23 plays a crucial role in various fields, including skeletal biology, mineral metabolism, and CKD.^30–33 Recent advancements indicate that reducing FGF23 signaling could mitigate these dental and skeletal anomalies, indicating the importance of understanding the regulatory mechanism of FGF23 production.^34–36 Therefore, it is crucial to develop a robust in vitro system to study FGF23-regulating mechanisms and identify potential therapeutic targets for FGF23 disorders. The monolayer culture of osteocytic and osteoblastic cell lines has been widely used for in vitro FGF23 studies. However, the monolayer culture lacks the hierarchical structure and cell–cell contact, which is crucial for the function of osteocytes and osteoblasts; thus, the FGF23 expression in these systems is extremely low or even undetectable.^37–39

In order to mimic the hierarchical structure within native tissue, numerous studies have explored 3D culture approaches including using scaffolding biomaterials or cell pallets generated by centrifugation or hanging drop techniques.^31–34 However, the complex and labor-intensive procedures of the aforementioned approaches limit their large-scale application. In this study, we have employed a high-throughput approach to generate stable osteoblastic spheroids composed of a nonadhesive microwell array prepared by our previously established soft photolithographic technique.¹⁶ Compared with the conventional 3D culture approach, our approach could generate versatile cellular spheroids composed of 200 cells in a high-throughput manner.^16,26 In addition, the submicron-scale size of the obtained cellular spheroids facilitates the nutrient diffusion during in vitro culture and ensures cell viability and homogenous differentiation within the spheroids.²⁶

In this study, the MC3T3E1 cell spheroids showed a significant upregulation of FGF23 after 24 h in both gene and protein levels, indicating the maturation of the osteoblastic cells.^40–42 Similar upregulation of FGF23 and Ocn was also observed in primary mouse periosteum derived skeletal stem/progenitor cell spheroids (Supplementary Fig. S1), suggesting the validity of this culture system on primary cells. We also explored the mechanism driving FGF23 upregulation. Our data revealed that the elevated FGF23 expression in MC3T3-E1 cell spheroids correlates highly with the elevated PTH/PTH1R signaling, which directly regulates osteoblast maturation and FGF23 production.⁴³ The exact mechanism driving the elevation of PTH/PTH1R signaling still needs further investigation. One possible explanation is that the mechanical cues within cell–cell and cell–matrix interactions thus subsequently enhance the endogenous ligand recognition and structural transition of a PTH receptor.⁴⁴ These findings indicate that targeting the PTH/PTH1R signaling pathway could be a potential therapeutic strategy for disorders involving abnormal FGF23 production. In light of this, our high-throughput method for generating osteoblastic spheroids offers significant advantages for target selection using small inhibitor libraries and CRISPR libraries.

In summary, in this study, we report a high-throughput approach to generate preosteoblastic cell spheroids with enhanced FGF23 production driven by the activation of PTH/PTH1R signaling. The scalability and efficiency of this approach allow for rapid screening and identification of potential therapeutic targets involved in FGF23 regulation. This capability provides a robust platform to explore and validate novel inhibitors or genetic modifications aimed at modulating FGF23 production. In addition, the reported high-throughput approach is not only limited to cell lines but also applicable to primary cells, which opens the possibility of screening the cells harvested from patients suffering from FGF23 disorders.

Footnotes

Acknowledgments

The authors would like to acknowledge Ms. Y. Zhang, Ms. Y. Zuo, and Mr. W. Yang for their great assistance in laboratory work. The authors would also like to appreciate the financial support from the National Natural Science Foundation of China, the Key Research and Development Program of Hubei Province, the Fundamental Research Funds for the Central Universities, and the ITI Research Foundation.

Author’s Contributions

J.J.: contributed to data acquisition and interpretation, performed all statistical analyses, and drafted and critically revised the article. J.Z., H.L., and S.J.: contributed to data acquisition and interpretation and critically revised the article. Q.H. and W.J.: contributed to conception, design, and data interpretation and drafted and critically revised the article.

Disclosure Statement

No competing financial interests exist.

Funding Information

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the National Natural Science Foundation of China (No. 82072483 to Q.H. and No.82170931 to W.J.); the Key Research and Development Program of Hubei Province (2022BCA052 to Q.H.); the Fundamental Research Funds for the Central Universities (No. 2042024YXB018 to W.J.); and the ITI research grant (No.1721–2022 to W.J.).

Supplementary Material

References

Bergwitz

, Jüppner

. Regulation of phosphate homeostasis by PTH, vitamin D, and FGF23. Annu Rev Med, 2010; 61(1):91–104; doi: 10.1146/annurev.med.051308.111339

Agoro

, White

. Regulation of FGF23 production and phosphate metabolism by bone–kidney interactions. Nat Rev Nephrol, 2023; 19(3):185–193; doi: 10.1038/s41581-022-00665-x

Edmonston

, Wolf

. FGF23 at the crossroads of phosphate, iron economy and erythropoiesis. Nat Rev Nephrol, 2020; 16(1):7–19; doi: 10.1038/s41581-019-0189-5

Yamazaki

, Okazaki

, Shibata

, et al. Increased circulatory level of biologically active full-length FGF-23 in patients with hypophosphatemic rickets/osteomalacia. J Clin Endocrinol Metab, 2002; 87(11):4957–4960; doi: 2020071616061761100

Trombetti

, Al-Daghri

, Brandi

, et al. Interdisciplinary management of FGF23-related phosphate wasting syndromes: A consensus statement on the evaluation, diagnosis and care of patients with X-linked hypophosphataemia. Nat Rev Endocrinol, 2022; 18(6):366–384; doi: 10.1038/s41574-022-00662-x

Edmonston

, Grabner

, Wolf

. FGF23 and klotho at the intersection of kidney and cardiovascular disease. Nat Rev Cardiol, 2024; 21(1):11–24; doi: 10.1038/s41569-023-00903-0

Dussold

, Gerber

, White

, et al. DMP1 prevents osteocyte alterations, FGF23 elevation and left ventricular hypertrophy in mice with chronic kidney disease. Bone Res, 2019; 7(1):12; doi: 10.1038/s41413-019-0051-1

Domazetovic

, Falsetti

, Ciuffi

, et al. Effect of oxidative stress-induced apoptosis on active FGF23 levels in MLO-Y4 cells: The protective role of 17-β-Estradiol. Int J Mol Sci, 2022; 23(4):2103; doi: 10.3390/ijms23042103

Courbon

, Francis

, Gerber

, et al. Lipocalin 2 stimulates bone fibroblast growth factor 23 production in chronic kidney disease. Bone Res, 2021; 9(1):35; doi: 10.1038/s41413-021-00154-0

10.

Hemmatian

, Jalali

, Semeins

, et al. Mechanical loading differentially affects osteocytes in fibulae from lactating mice compared to osteocytes in virgin mice: Possible role for lacuna size. Calcif Tissue Int, 2018; 103(6):675–685; doi: 10.1007/s00223-018-0463-8

11.

Klein-Nulend

, van Oers

RFM

, Bakker

, et al. Bone cell mechanosensitivity, estrogen deficiency, and osteoporosis. J Biomech, 2015; 48(5):855–865; doi: 10.1016/j.jbiomech.2014.12.007

12.

Sun

, Choudhary

, Mannion

, et al. Ex vivo replication of phenotypic functions of osteocytes through biomimetic 3D bone tissue construction. Bone, 2018; 106:148–155; doi: 10.1016/j.bone.2017.10.019

13.

Wheeler

, Clinkenbeard

. Regulation of fibroblast growth factor 23 by iron, EPO, and HIF. Curr Mol Biol Rep, 2019; 5(1):8–17; doi: 10.1007/s40610-019-0110-9

14.

David

, Martin

, Isakova

, et al. Inflammation and functional iron deficiency regulate fibroblast growth factor 23 production. Kidney Int, 2016; 89(1):135–146; doi: 10.1038/ki.2015.290

15.

Fontoura

, Viezzer

, dos Santos

, et al. Comparison of 2D and 3D cell culture models for cell growth, gene expression and drug resistance. Mater Sci Eng C Mater Biol Appl, 2020; 107:110264; doi: 10.1016/j.msec.2019.110264

16.

Leijten

, Teixeira

LSM

, Bolander

, et al. Bioinspired seeding of biomaterials using three dimensional microtissues induces chondrogenic stem cell differentiation and cartilage formation under growth factor free conditions. Sci Rep, 2016; 6(1):36011; doi: 10.1038/srep36011

17.

Sawa

, Fujimoto

, Sawa

, et al. Alternating differentiation and dedifferentiation between mature osteoblasts and osteocytes. Sci Rep, 2019; 9(1):13842; doi: 10.1038/s41598-019-50236-7

18.

Thompson

, Uzer

, Brobst

, et al. Osteocyte specific responses to soluble and mechanical stimuli in a stem cell derived culture model. Sci Rep, 2015; 5(1):11049; doi: 10.1038/srep11049

19.

Zhang

, Bakker

, Klein-Nulend

, et al. Studies on osteocytes in their 3D native matrix versus 2D in vitro models. Curr Osteoporos Rep, 2019; 17(4):207–216; doi: 10.1007/s11914-019-00521-1

20.

Choudhary

, Ramasundaram

, Dziopa

, et al. Human ex vivo 3D bone model recapitulates osteocyte response to metastatic prostate cancer. Sci Rep, 2018; 8(1):17975; doi: 10.1038/s41598-018-36424-x

21.

Belardi

, Son

, Felce

, et al. Cell–cell interfaces as specialized compartments directing cell function. Nat Rev Mol Cell Biol, 2020; 21(12):750–764; doi: 10.1038/s41580-020-00298-7

22.

Kim

, Adachi

. Cell-fate decision of mesenchymal stem cells toward osteocyte differentiation is committed by spheroid culture. Sci Rep, 2021; 11(1):13204; doi: 10.1038/s41598-021-92607-z

23.

Bhattacharjee

, Zehnder

, Rowe

, et al. Writing in the granular gel medium. Sci Adv, 2015; 1(8):e1500655; doi: 10.1126/sciadv.1500655

24.

Lee

S-Y

, Koo

I-S

, Hwang

, et al. In Vitro three-dimensional (3D) cell culture tools for spheroid and organoid models. SLAS Discov, 2023; 28(4):119–137; doi: 10.1016/j.slasd.2023.03.006

25.

Atala

, Kasper

, Mikos

. Engineering complex tissues. Sci Transl Med, 2012; 4(160):160rv12; doi: 10.1126/scitranslmed.3004890

26.

Zhu

, Zhang

, Jin

, et al. Endochondral repair of jawbone defects using periosteal cell spheroids. J Dent Res, 2024; 103(1):31–41; doi: 10.1177/00220345231205273

27.

Chijiwa

, Mishima

, Hagiwara

, et al. Inhibition of forskolin-induced neurite outgrowth and protein phosphorylation by a newly synthesized selective inhibitor of cyclic AMP-dependent protein kinase, N-[2-(p-bromocinnamylamino)ethyl]-5-isoquinolinesulfonamide (H-89), of PC12D pheochromocytoma cells. Journal of Biological Chemistry, 1990; 265(9):5267–5272; doi: 10.1016/S0021-9258(19)34116-X

28.

Langmead

, Salzberg

. Fast gapped-read alignment with Bowtie 2. Nat Methods, 2012; 9(4):357–359; doi: 10.1038/nmeth.1923

29.

Chen

, Wu

, Li

, et al. TBtools-II: A “one for all, all for one” bioinformatics platform for biological big-data mining. Mol Plant, 2023; 16(11):1733–1742; doi: 10.1016/j.molp.2023.09.010

30.

Donate-Correa

, Muros-de-Fuentes

, Mora-Fernández

, et al. FGF23/Klotho axis: Phosphorus, mineral metabolism and beyond. Cytokine Growth Factor Rev, 2012; 23(1–2):37–46; doi: 10.1016/j.cytogfr.2012.01.004

31.

Sun

, Yu

. FGF23 actions in CKD-MBD and other organs during CKD. Curr Med Chem, 2023; 30(7):841–856; doi: 10.2174/0929867329666220627122733

32.

, Zhang

, Wang

, et al. Endocrine regulation of extra-skeletal organs by bone-derived secreted protein and the effect of mechanical stimulation. Front Cell Dev Biol, 2021; 9:778015; doi: 10.3389/fcell.2021.778015

33.

Fukumoto

, Martin

. Bone as an endocrine organ. Trends Endocrinol Metab, 2009; 20(5):230–236; doi: 10.1016/j.tem.2009.02.001

34.

Lira dos Santos

, Nakajima

, Po

, et al. Dental impact of anti-fibroblast growth factor 23 therapy in X-linked hypophosphatemia. Int J Oral Sci, 2023; 15(1):53; doi: 10.1038/s41368-023-00259-8

35.

Xiao

, Liu

S-H

, et al. Novel small molecule fibroblast growth factor 23 inhibitors increase serum phosphate and improve skeletal abnormalities in Hyp mice. Mol Pharmacol, 2021; 101(6):408–421; doi: 10.1124/molpharm.121.000471

36.

Fukumoto

. Recent basic and clinical findings concerning FGF23: A bone-derived hormone. Current Opinion in Endocrine and Metabolic Research, 2023; 29:100445; doi: 10.1016/j.coemr.2023.100445

37.

Sun

, Choudhary

, Mannion

, et al. Ex vivo construction of human primary 3D–networked osteocytes. Bone, 2017; 105:245–252; doi: 10.1016/j.bone.2017.09.012

38.

Sun

, Gu

, Zhang

, et al. Ex vivo 3D osteocyte network construction with primary murine bone cells. Bone Res, 2015; 3(1):15026; doi: 10.1038/boneres.2015.26

39.

Knowles

, Chanalaris

, Koutsikouni

, et al. Mature primary human osteocytes in mini organotypic cultures secrete FGF23 and PTH1-34-regulated sclerostin. Front Endocrinol (Lausanne), 2023; 14:1167734; doi: 10.3389/fendo.2023.1167734

40.

Michigami

. Paracrine and endocrine functions of osteocytes. Clin Pediatr Endocrinol, 2023; 32(1):1–10; doi: 10.1297/cpe.2022-0053

41.

Mironov

, Haque

, Atfi

, et al. Phosphate dysregulation and metabolic syndrome. Nutrients, 2022; 14(21):4477; doi: 10.3390/nu14214477

42.

Guo

Y-C

, Yuan

. Fibroblast growth factor 23 and bone mineralisation. Int J Oral Sci, 2015; 7(1):8–13; doi: 10.1038/ijos.2015.1

43.

Coricor

, Serra

. TGF-β regulates phosphorylation and stabilization of Sox9 protein in chondrocytes through p38 and Smad dependent mechanisms. Sci Rep, 2016; 6:38616; doi: 10.1038/srep38616

44.

Kobayashi

, Kawakami

, Kusakizako

, et al. Endogenous ligand recognition and structural transition of a human PTH receptor. Mol Cell, 2022; 82(18):3468–3483.e5; doi: 10.1016/j.molcel.2022.07.003

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.37 MB

0.00 MB